On the Design and Hardware Architectures for Lossless Data Compression

• Main Authors: Liang-Ying Liu, 劉良潁
• Other Authors: Jhing-Fa Wang, Jau-Yien Lee
• Format: Others
• Language: en_US
• Published: 1995
• Online Access: http://ndltd.ncl.edu.tw/handle/34179312616245238661

博士 === 國立成功大學 === 電機工程研究所 === 84 === Lossless data compression is a useful strategy to the communication system and the storage system. Demands on the real-time lossless compression, such as the HDTV system and the multi-medium system, have been greatly increased in recent years. The Huffman coding is one of widely-used statistic coding scheme. In spite amount of researches devoted on the hardware design for Huffman coding, few of them addressed the adaptive designs. The adaptive Huffman coding is expert on the source with unknown probabilities of symbols. It is important to the compression of the general source and worth efforts on its hardware designs. In this dissertation, contributions to the design methodology and the architectures for the adaptive Huffman coding are presented. They can be listed in following topics. (1) A DFG reforming/scheduling algorithm and a fast estimation technique. (2) The VLSI implementation for the dynamic Huffman code. (3) The concurrent algorithm and presetting method for the ynamic Huffman coding. (4) Architectures for implementing the concurrent dynamicuffman coding. (5) A modified Huffman code. This code has a significantdvantage on hardware cost. (6) The Lempel-Ziv encoder and decoder designs. The dynamicuffman encoder is included in this encoder. The betterompression efficiency can be achieved.